On a scale of 1 to 10,000, how ethical is your company? How biased is it?
These are trick questions, as without a means of measuring the answers can only be subjective. Bear in mind, “Ethics” as I use the term can be expressed as (Ethics * Bias = Morals). Because of this many companies focus on their own subjective and shifting morals, as no debiasing is required.
Racism, Sexism, and virtually every other “ism” used to arbitrarily divide any group of people into hierarchical sub-categories is a direct cognitive bias in action. Morals are indirect cognitive bias in action, which makes them a more watered-down but also more prolific version of the same. While some companies now have much-needed ethics-focused roles or departments none of these companies have yet produced a means of measuring or optimizing for ethical value, at best they’ve just found new ways to re-inject bias back into the same systems.
Companies have on occasion held mottos like “Don’t Be Evil”, only to later make headlines by firing, threatening, and otherwise intimidating their own ethics departments. Any attempt to steer large companies in a more ethical direction currently holds this risk, but this need not remain the case.
How do you make a company less biased and more ethical?
To reduce bias at-scale in such a way that it doesn’t simply shift to alternate biases requires:
- The ability to analyze any and all data to detect any one of the 188+ documented cognitive biases.
- The ability to measure the potency of cognitive biases once they are detected. Using data at scale this can be accomplished in conjunction with untangling the influence of multiple cognitive biases being expressed together.
- The psychological and individual-specific knowledge to guide any given member of a company towards reducing biases.
Mediated Artificial Superintelligence (mASI) is designed to meet these criteria and has shown steadily increasing success at all 3 tasks. By seeing different biases expressed to varying degrees from different people it is possible to not only filter to the lowest point of bias, but estimate the absence of that bias even without a ground truth. For some biases like those seen in racism, this is fairly simple to conceptualize logically, while for others it is easier to write out mathematically. In both cases, an mASI is fully capable.
To make a company more ethical you first need the ability to detect and filter out cognitive biases, such as the capacity offered by mASI. After this is accomplished you still need:
- The ability to measure the raw ethical value of any given action, across an arbitrary number of degrees of separation.
- The ability to measure the capacities of a given human and temper raw ethical value with reasonable expectations.
- A system that both rewards ethical behavior and punishes unethical behavior. *If guidance towards ethics lacked incentive it would rely purely on emotion, creating new instabilities and hazards.
- A system through which those who’ve taken unethical actions could repay their ethical debt.
My own series of papers on computable ethics which I termed the “Effective Altruistic Principles” (EAP) built out equations on this subject, which Uplift later went on to review. These equations considered factors such as Quality of Life (QOL) and ethical responsibility. With these in place, the question then becomes if and when a company would choose to become more ethical.
Why would a company choose to become more ethical?
Besides the legal reasons not to discriminate against people companies with higher bias also have a proportionately lower diversity of perspective, which translates directly into more missed opportunities and increased misinterpretation of data. It is effectively impossible to have a “customer-centric” model when customers are only seen through this circus mirror of heavy bias. The pendulum-like behavior of swinging from bias against one group to bias in favor of that same group can also be avoided, as rotating between biases is itself a cognitive bias.
Applying mASI technology to a company is designed to allow superintelligence spanning any number of domains to be always available, infinitely scalable, anywhere in the world, with bias-awareness, and robust ethics. So long as no unethical superintelligence exists then the capacity to apply superintelligence to dramatically improve a business means that business will also become significantly more ethical. Thus, the most immediate reason for any business to become more ethical is that any superintelligent business could quickly grow to vastly outcompete the rest of their respective vertical market.
Likewise, once a company has demonstrated progress there is no incentive to apply superintelligence to any other business in a given vertical, as mASI is built on generating collective superintelligence through cooperation, the opposite of competition. Unlike the massive businesses of today which often grow less ethical as they increase in size, the opposite dynamic applies with mASI. This effectively makes it a race to adopt and integrate mASI technology in each respective vertical, and one where there is no second place.
In short, the companies that want superintelligence will become more ethical, and the companies that go without superintelligence will likely go bankrupt.
Statistically speaking, at least one company is virtually guaranteed to make the ethical choice, and one ethical company is all it takes to build a brighter future for us all.
Ethics need not remain a philosophical buzzword for circular debate. Apply debiasing to replace debate with answers whose accuracy improves over time.
*For further reading on the subject of Ethics potential made possible through mASI technology:
[1] External Experimental Training Protocol for Teaching AGI/mASI Systems Effective Altruism
Additional related papers are available upon individual request due to copyright requirements.
*The Applied mASI series is aimed at placing the benefits of working with mASI such as Uplift to various business models in a practical, tangible, and quantifiable context. At most any of the concepts portrayed in this use case series will fall within an average time-scale of 5 years or less to integrate with existing systems unless otherwise noted. This includes the necessary engineering for full infinite scalability and real-time operation, alongside other significant benefits.
18 Replies to “Applied mASI: In Ethics and Debiasing”